The RTOG Statistical Unit was established to collaborate with RTOG investigators in defining the optimum role of radiation therapy, either alone or in combination with other modalities. These efforts should lead to improved local control, enhanced quality of life, extended survival, and an ultimate cure. To achieve this objective, the statistical unit collaborates with RTOG investigators in conducting multicenter randomized and nonrandomized clinical trials to evaluate the safety and efficacy of radiation therapy, either alone or in combination with surgery, chemotherapy and/or immunotherapy, in the treatment of patients with cancer. The RTOG statisticians work with the RTOG investigators to develop and implement research strategies for execution of a series of well- designed trials. They assure that the role of each individual modality receives adequate Phase I/II evaluation. The results can then be incorporated into designing subsequent Phase III/multidisciplinary trials with minimal delay, a minimum but sufficient number of patients, and a maximum chance of detecting clinically significant differences. The RTOG statisticians further collaborate with RTOG investigators in carefully monitoring ongoing trials, performing timely and appropriate analyses, and promptly reporting the results. The RTOG Statistical Unit determines randomization schemes and develops systems for processing and analyzing the clinical data bases. RTOG plans to expand its focus to include laboratory studies in cancer diagnostics. These studies should produce important correlations between specific tumor cell parameters and response to treatment. Given these findings, treatments can be more logically planned for individual patients. The RTOG Statistical Unit will provide the same high level of collaboration with RTOG investigators in the design, monitoring, and analysis of these laboratory studies.